🔒 Synchronization

Threads share memory, which means two threads can read/write the same data at the same time.

Synchronization primitives help you coordinate access and avoid corruption.

---

✅ Locks

A Lock protects a critical section:

import threading

lock = threading.Lock()

with lock:
    # only one thread at a time here
    shared_counter += 1

Use the smallest critical section possible.

---

✅ Events

An Event is a simple signal:

• one thread waits: event.wait()
• another thread signals: event.set()

---

✅ Semaphores

A Semaphore limits how many threads may enter a section concurrently.

---

✅ Prefer Message Passing

Often the cleanest approach is to avoid sharing state:

• use queue.Queue to pass work/results between threads
• let each thread own its local state

---

🔍 Key Takeaways

• Use Lock when you must share state.
• Use Event for one-way signaling.
• Use Queue to reduce shared-state complexity.

---

Back: Module 12 README